The extent of incorporation of excess deuterium in the platinum

The extent of incorporation of excess deuterium in the platinum-catalyzed reduction of unsubstituted cycloolefins by H2 in D2O/THF increases with the ...
0 downloads 0 Views 291KB Size
J . Am. Chem. SOC. 1991, 113, 369-310

the platinum centere3 (ii) The reduction of ( I ,5-cycloo~tadiene)Pt(C~H ((COD)PtPh2)-a ~)~ complex in which transfer with inversion at carbon is not possible-proceeds at a rate comparable to that of the reductions of (COD)PtR, (R = Me, Et, n-Pr, i-Pr, i-Bu), complexes in which transfer with inversion at carbon is possible.2 (iii) The reduction of (COD)PtNp2-a complex that generates neopentyl. m~ieties'~-also proceeds at a rate comparable to that of the reductions of the (COD)PtR,

369 dav= 7.5

2ou 40

0

d,, = 7.6

20

We conclude that the reduction of exo-2-norbornyL and endo-2-norbornyl. moieties occurs with predominant retention of stereochemistry. The result from the reduction of exo-2-norbornyl. appears to be stereochemically straightforward, with the only competing reaction being activation of ca. 20% of the cis C(3)-H (exo) bonds (probably by @-hydrideelimination). The reduction of endo-2-norbornyl. is more complicated. The observed results are accounted for by a model involving ca. 35% isomerization of endo-2-norbornyL to exo-2-norbornyl., but with the conversion of C t to CD again occurring with retention of stereochemistry. We have confirmed the structures of 1 and 2 using X-ray crystallography and have characterized (by MS and ' H NMR) the alkane products of the reductions of 1 and 2. These details, and their interpretations, will be described separately.

6oi Cvcloheatanes

20

(17) Displacement with inversion a t neopentyl centers is typically much slower than at other alkyl centers. See, for example: Ingold, C. K. Q.Reu., Chem. Sac. 1977, 1 1 , 1-14. 20

601 CvcloDentadecanes ,,8,52=[)& , , ,

The Extent of Incorporation of Excess Deuterium in the Platinum-Catalyzed Reduction of Unsubstituted Cycloolefins by H, in D,O/THF Increases with the Strain Energy of the Product Cycloalkanes' T. Randall Lee and George M. Whitesides* Department of Chemistry, Harvard University Cambridge, Massachusetts 02138 Received August 9, 1990 This paper describes the isotopic compositions of cycloalkanes produced by the catalytic reduction of cycloolefins (C,H2,2) using H2 and platinum black in l:i v:v T H F / D 2 0 , D , I (eq We

-

H2 (PH2 0.17 a m ) , PI

cycloolefin

D 2 0 ( D + ) / T H F , T = 75

c

cycloalkane-d,

(1 )

O C

conducted the reductions under conditions in which the rate of mass transport of H2to the surface of the catalyst limited the rate of r e a ~ t i o n . ~ , ~Under these conditions, H. on the surface equilibratcs cssentially completely with D 2 0 in solution. Figure 1 summarizes the isotopic compositions of the cycloalkane~.~The ( I ) The National Science Foundation (Grant CHE-88-12709), the Office of Naval Research, and the Defense Advanced Research Projects Agency supported this work. (2) We chose D,O/THF ( I : ] v:v; P D D ~ = O I ) because a survey of solvent systems indicated that this system yielded cyclooctanes having the highest content of deuterium in reductions of cyclooctene. (3) We performed the reductions in a specially designed round-bottomed flask (volume = 225 mL): a small teat (volume ca. IO mL) protruded from the bottom of the flask; a Swagelok valve and a rubber septum capped the flask. After charging the teat with a stir bar ( 3 / 8 X 3/16 in.) and 40 mg of Pt black, we purged the flask with Ar and added 0.5 mL of dry T H F and 0.5 mL of D 2 0 (adjusted to pD = 1 with D2S04). The apparatus was purged with a mixture of 10% H2 in Ar for 1 min with stirring; the rate of rotation of the stir bar was 1400 rpm. We pressurized the vessel to I O psi ( P H I= 0.17 atm), stirred for 4 min, placed the flask in an oil bath (75 i 3 "C), and stirred for 5 min. We removed the solvent via cannula and added the substrate ( I 5-20 mg in 3 mL of D 2 0 / T H F ) via syringe. Stirring was started and continued for I h. Analysis by G C / M S indicated that all reductions were complete. (4) Miller, T. M.; Izumi, A. N.; Shih, Y.-S.;Whitesides, G. M. J . Am. Chem. Sac. 1988, 110, 3 146-3 156.

0002-7863/91/15 I3-369$02.50/0

0

, , ,

,

, , , C , , ,

,

,

,

,

,A

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Number of Deuterium Figure 1. Isotopic distributions and values of d,, for the cycloalkanes from the reductions of the corresponding cycloolefins by H 2 and D20/ T H F ( I : I v:v; = 1) over platinum black.5 T h e difference in values of d,, from duplicate runs was, in all cases, 50.2. 1 n. I."

0.90.8d,, d2n

0.70.6-

-5.0

0 2 4 6 8 10 12 14 16 18 20 Strain Energy (kcalimol)

Figure 2. Values of d,,/d,, vs strain energies of cycloalkanes (C,H2n).5J7 For consistency, the strain energies a r e all calculated values, obtained using Macromodel V2.0.'8319 T h e line drawn through the points was determined by linear regression.

important observations are as follows: ( I ) the isotope of hydrogen found in the cycloalkanes was predominantly that present in D 2 0 rather than that present in H,; (2) the cycloalkanes contained deuterium in large excess of that required for stoichiometric (5) Isotopic compositions were determined by G C / M S 6 The average content of deuterium, d,,, describes the extent of incorporation of deuterium. Isotopic abundances were corrected for natural abundance of I T . Values of d,, are probably accurate to i 0 . 3 D. m

d,, =

En('% alkane-d,) n- I

(6) Miller, T . M.; McCarthy, T. J.; Whitesides, G.M. J . Am. Chem. Sac. 1988, / I O , 3 I 56-3 163.

0 1991 American Chemical Society

J . Am. Chem. SOC.1991, 113, 370-372

370

reduction of the double bond (approaching complete exchange of D for H in some cases); and (3) each substrate gave the perdeuterated isotopomer as the major product. The observation that extensively polydeuterated cycloalkanes can be produced from cycloolefins with Hz as the reductant is remarkable and potentially useful in the synthesis of isotopically labeled compounds. The ability to incorporate deuterium into organic compounds by the metal-catalyzed reduction of olefins with H2 and D20/THF avoids the need for D2. In addition, this technique should be applicable to the incorporation of tritium into organic compounds. Previous studies of the hydrogenation of olefins in protic solvents showed that surface alkyls on platinum rapidly undergo isotopic exchange of hydrogen with H. (D,) and that the product alkanes Our contain the isotope from the solvent (ROH or results demonstrate that by working under mass-transport-limited conditions, isotopic exchange can be made more extensive than in these earlier reports. The exchange undoubtedly requires exchange between HI (D.) and the OH (OD) group of the solvent (eq 2). The mechanism of the exchange shown in eq 2 is not well established, although several mechanisms have been proposed.I*I6 H. ROD F? D* + ROH (2) Because the exchange of H / D between the surface of platinum and the protic solvent is, under the conditions used here, faster than the conversion of surface alkyls to alkanes, the content of deuterium in the product alkanes provides a valuable mechanistic tool: a measure of the rate of exchange (and thus of C-H bond activation) of the surface alkyls relative to the rate of their reductive elimination as alkanes. Comparison of the ratio dav/d2, from the reduction of a series of cycloalkenes demonstrates the utility of this approach (Figure 2). These data indicate that the rates of C-H activation (relative to the rates of reductive elimination) of surface cycloalkyls decrease from cyclodecyl, to cyclohexyl.. The value of d,,/d2, correlates with the strain energy of the product cycloalkanes ( r = 0.95).17-19Thus, as the cycloalkane is varied from one having low strain energy to one having high strain energy, the transition state for reductive elimination increases in energy relative to that for isotopic exchange. The two simple limiting hypotheses rationalizing this inference are that strain is relieved in the transition state for isotopic exchange (C-H activation) or that strain is concentrated in the transition state for reductive elimination?1 We emphasize that AG* for reductive

+

(7) Eidinoff, M. L.; Knoll, J. E.; Fukushima, D. K.; Gallagher, T. F. J . Am. Chem. SOC.1952, 74, 5280-5284. (8) Phillipson, J. J.; Burwell, R. L., Jr. J . Am. Chem. SOC.1970, 92, 61 25-61 33. (9) Trahanovsky, W. S.; Bohlen, D. H. J . Org. Chem. 1972, 37, 21 92-21 97. (IO) Shimazu, K.; Kita, H. J. Chem. Soc., Faraday Trans. I 1985, 81, 175-183. ( I I ) Fujikawa, K.; Kita, H.; Sato, S. J . Chem. Soc., Faraday Trans. 1 1981, 77, 3055-3071. (12) Fujikawa, K.; Kita, H.; Miyahara, K.; Sato, S.J . Chem. Sor., Faraday Trans. I 1975, 71, 1573-1581. (13) Kita, H. fsr. J . Chem. 1979, 18, 152-161. (14) Sagert, N. H.; Pouteau, R. M. L. Can. J . Chem. 1974,52,2960-2967. (15) Miyamoto, S.-I.;Sakka, T.; Iwasaki, M. Can. J . Chem. 1989, 67, 857-861. (16) Rolston, J . H.; Goodale, J . W. Can J . Chem. 1972, 50, 1900-1906. (17) The maximum possible content of deuterium in a cycloalkane is d2". ( I 8) We calculated strain energies for the lowest energy conformations of the cycloalkanes given in the following: Burkert, U.; Allinger, N. L. Molecular Mechanics; ACS Monograph 177; American Chemical Society: Washington, DC 1982; pp 89-108 (for C5-C12). Anet, F. A. L.; Rawdah, T. N. J . Am. Chem. SOC.1978, 100, 7810-7814 (for C15).The strain energies for these conformations were calculated with Macromodel V2.0 using the MM2(85) parameter set: Still, W. C.; Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.;Hendrickson, T. MacroModel V2.0; Dept. of Chemistry, Columbia University: New York, NY (19) Thc strain cncrgies shown in Figure 2 were obtained by standard modification of the values given by force field calculations (SE = SEHM2 SEatru,nlcas).20 The values for SEllrBlnlCBB were empirically determined to be n X 0.65 kcal/mol for a given cycloalkane (C,H2#). (20) See, for examplc: DeTar, D. F.; Tenpas, C. J . J . Org. Chem. 1976, 41, 2009-2013.

elimination by reaction of R. with H2 is dominated by the free energy term desciibing generation of H. by a mass-transportlimited process. We therefore hypothesize that the correlation in Figure 2 reflects primarily structure-dependent differences in rates of C-H activation. The heterogeneous polydeuteration of hydrocarbons probably involves some combination of aa-activation, a@-activation,and T-allyl formation.22 The differences in strain energies between the transition states leading to these intermediates and the transition states leading to the cycloalkane products should correlate with dav/d2,,.Calculating strain energies for the intermediates is, however, too complicated to be approached through realistic models of surface species.23 To date, our efforts to obtain a correlation between the experimental values of da,/d2, and relative strain energies calculated for simple models of surface intermediates have not been successful. Acknowledgment. We thank our colleagues John Burns and Watson Lees for assistance with the force field calculations. ~~~~~

(21) There are several reports in the literature of correlations between the strain energy and the rates of reaction of functionalized cycloalkanes. For recent examoles. see: Schneider. H. J.: Thomas. F. J . Am. Chem. SOC.1980. 102, 1424-1'425: Schneider, H.'J.; Schmidt, 6.;Thomas, F. J . Am. Chem. SOC. 1983, 105, 3556-3563. (22) Lebrilla, C. B.; Maier, W. F. J . Am. Chem. SOC.1986, 108, 1606-1 6 16 and references therein. (23) The hydrocarbon species believed to exist on the surfaces of noble metal catalysts were summarized recently: Cogen, J. M.; Maier, W. F. J . Am. Chem. SOC.1986, 108, 7752-7762 and ref 22 of this paper.

Four-Dimensional [13C,1H,13C,1H] HMQC-NOE-HMQC NMR Spectroscopy: Resolving Tertiary NOE Distance Constraints in the Spectra of Larger Proteins Erik R. P. Zuiderweg,* Andrew M. Petros, Stephen W. Fesik, and Edward T. Olejniczak Pharmaceutical Discovery Division. Abbott Laboratories D47G, AP9, Abbott Park, Illinois 60064 Received September 27, I990 Three-dimensional (3D) heteronuclear-resolved nuclear Overhauser effect (NOE) NMR spectroscopy is a powerful tool to improve the resolution of NOE spectra of medium-sized proteins (MW